﻿read_me.txt

Data and Program Description for RESTAT article 
“The Impact of Price Discrimination on Revenue: Evidence from the Concert Industry”, 
by Pascal Courty and Mario Pagliero

******************************************************************
IMPORTANT NOTE
******************************************************************

The dataset used in this paper includes concert data and demographic data. Concert data are not freely available but can be easily obtained from Billboard (http://www.billboard.com). The data includes concert level information for the “top 100 grossing artists over the period 1992-2005”. The names of the artists included in our dataset are reported in the file artists.txt. In some instances, two artists perform together and we treat couples of artists as separate artists.

Demographic variables at the city or place level are obtained from the 2000 Census (http://www.census.gov/main/www/cen2000.html). Concert data and Census data are merged using the city variable in the concert data and the city or place variable in the Census data. We matched locations at the most disaggregate possible level. 
In order to ease the replication of the results the demo.dta file contains the demographic variables 
for matched cities. Hence, the concert data obtained from Billboard can be merged with demographic variables in demo.dta (for example using the "merge" command in stata). Stata 10se was used in the analysis.

The file concert.do describes the code for creating the variables necessary for the analysis and computing the results. 

List and description of concert-level variables:
artist1= name of the artist 
startyear= year in which the concert takes place 
state =state in which the concert takes place
country= country in which the concert takes place (we use U.S. data only)
city= city in which the concert takes place 
venue = name of the venue in which the concert takes place
ticketprice1 =highest price
ticketprice2 =2nd highest price
ticketprice3 =3rd highest price
ticketprice4 =4th highest price
usgross=concert revenue 
attendance =total concert attendance
capacity = total available capacity
num_prices =number of price categories used
promoter1 =name of promoter
tour = numerical indicator for the tour
 
List and description of demographic variables:
population2000= 2000 Census population
pop_frac_white2000= proportion of white population
pop_frac_black2000= proportion of black population
e_tot = total employment
e_m_management_001_359 = male emplyment in managerial occupations
e_m_services_360_469 = male employment in service industry
e_m_sales_470_599 =  male employment in sales
e_m_farming_600_619 = male employment in farming
e_m_construction_620_769= male employment in construction
e_m_production_770_979 =male employment in production
Female emplyment is similarly defined:
e_f_management_001_359
e_f_services_360_469
e_f_sales_470_599
e_f_farming_600_619
e_f_construction_620_769
e_f_production_770_979
h_tot_inc= total hausehold income
Variables h__10k, h_10_14k, h_15_19k, h_20_24k, h_25_29k, h_30_34k, h_35_39k, h_40_44k, h_45_49k, h_50_59k, h_60_74k, h_75_99k, h_100_124k, h_125_149k, h_150_199k, h__200k
describe hausehold income by income barkets (e.g., 10k-14k). 
med_h_inc =median hausehold income






